Learning Consensus for Nonlinear Multi-Agent Systems with Iteration-Switching Topologies

In order to deal with uncertain nonlinear multi-agent systems with nonrepeatable mismatched input disturbance, an iterative learning controller is proposed in this paper. The main structure of this controller is constructed by a neural networks learning component and a robust learning component. The neural networks used as an approximator is designed to compensate for the system's nonlinearity. The robust learning component is used to overcome the nonlinear input gain and input disturbance. An adaptive law combining time- and iteration- domain is used to tune the controller parameters. The consensus convergence is proved by the composite energy function method. A numerical simulation is given to verify the effectiveness of the proposed control scheme.